Skip to main content

Full text of "On Some Species of Heliconia"

See other formats


Early Journal Content on JSTOR, Free to Anyone in the World 

This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in 
the world by JSTOR. 

Known as the Early Journal Content, this set of works include research articles, news, letters, and other 
writings published in more than 200 of the oldest leading academic journals. The works date from the 
mid-seventeenth to the early twentieth centuries. 

We encourage people to read and share the Early Journal Content openly and to tell others that this 
resource exists. People may post this content online or redistribute in any way for non-commercial 
purposes. 

Read more about Early Journal Content at http://about.jstor.org/participate-jstor/individuals/early- 
journal-content . 



JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people 
discover, use, and build upon a wide range of content through a powerful research and teaching 
platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit 
organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please 
contact support@jstor.org. 




V 



o 




Bull. Torrey Club, 30. 



PL. 30 




HELICONIA CHAMPNEIANA 



Vol. 30 No. 12 

BULLETIN 



OF THE 



TORREY BOTANICAL CLUB 



DECEMBER, 1903 



On Some Species of Heliconia 

By Robert F. Griggs 
(With Plates 29 and 30) 

Heliconia occupies in America the place in plant society that is 
filled by the banana and its congeners in the Old World tropics. 
Its anatomy is similar to that of its better-known relative. Like 
the banana it is a very difficult subject for the botanical collector. 
The parts are both thick and succulent and of such size as to 
make it impossible to represent the plant at all adequately on an 
herbarium sheet. As knowledge of plants from remote regions 
has been largely dependent on herbarium specimens, it is not diffi- 
cult to appreciate that Heliconia has been an unsatisfactory genus 
to work on. Only once before the present study was undertaken 
has it been studied to any extent in the field. This fact accounts 
for most of the innovations suggested in the present paper. 

Although the genus has been monographed several times 
within recent years, as far as practicable recourse was had to orig- 
inal literature in determination and comparison. Appended to the 
paper is a list of the papers consulted. It includes the original 
descriptions of twenty-six of the twenty-nine recognized species. 
Though no pretension is made to bibliographic completeness, it 
may serve as the beginning of a guide to anyone who desires to 
take up the study of the genus, as it is believed that with a few 
exceptions most of the important literature is cited. 

All our plants, except the Porto Rican Heliconia Borinquena, 
came from a region in eastern Guatemala, which centers in Messrs. 
Owen and Champney's coffee fine a Sepacuite, a few miles north of 

[The preceding number of the Bulletin, Vol. 30, No. 11, for November, 1903 
(30 : 571-640, pi. 23-28), was issued 4 N 1903.] 

641 



642 Griggs : Species of Heliconia 

Senaju. From there trips were made northeastward to Cahabon 
and in a more easterly direction up the valley of the Oxec, a good- 
sized creek emptying into the Cahabon river ten miles below the town. 
In addition, the most abundant species at least were collected along 
the road from Panzos to Sepacuite. We were informed that the 
region had never been visited by a botanist, the nearest that one had 
been being Senaju, where Captain John Donnell Smith made a short 
stop. The mountains about Sepacuite, which is about 3,000 feet 
above sea-level, are heavily forested and in nearly their primeval 
condition. Here one species, H. tortuosa, is very abundant. Few 
of the species, however, are forest plants. Around the old Indian 
town of Cahabon the country has been reduced to an almost tree- 
less, grass-grown desert, and on the edges of this desert where the 
woods are not very thick and yet the water-supply is sufficient, 
most of the species were found. Along the road from Panzos to 
Sepacuite much the same conditions prevail as around Cahabon. 

We wish to express our thanks to the owners of Sepacuite, to 
Mrs. Owen especially, and to Mr. Alfred Rock, of Setzimaj, for their 
kind hospitality and for the very great assistance they so eagerly 
offered us, without which it would have been impossible to make 
our collections. I am very grateful to Mr. John Donnell Smith, 
of Baltimore, for the loan of all of his material on the genus, and 
to Dr. W. A. Kellerman, of the Ohio State University, for valuable 
criticism and suggestions. Thanks are also extended to the officers 
of the National Herbarium for the loan of specimens. 

The references after the descriptions, c. g. y " no. 258 '," are to 
the collectors' numbers accompanying the sheets. Except where 
otherwise stated, all the plants are from a collection made by Mr. 

0. F. Cook and myself during March and April, 1902. This col- 
lection was made for the United States Department of Agriculture 
and is deposited in the herbarium of the department in the National 
Museum. The numbers of photographs cited are either serial or 
field numbers of the collection in the Office of Tropical Agricul- 
ture of the Department. 

Key to the Groups and Species Discussed in this Paper 

1. Leaves borne at intervals along an elongated stem, /. e., with the habit of a Zingiber ; 

mostly small (//. psittacoruni) . Subgenus Stenochlamys Baker. 

A. Leaves narrowly oblong, sessile, almost clasping the stem ; peduncle short or 

absent; branch -bracts red, narrow. //. Choconiana Wats. 



Griggs : Species of Heliconia 643 

B. Leaves lanceolate or ovate-lanceolate, rounded or subcordate at the base; 

mostly petioled; peduncle at least as long as the rachis ; branch-bracts yellow 

with red cheeks, broad, often subovate. //. crassa. 

II. Leaves all borne from nearly the same point on the stem, as in II. Borinquena (/. 

j) ; mostly very large. 

A. Branch-bracts more or less spaced out on the rachis, not overlapping, except 

at the base (//. Bi/iai). Subgenus Platychlamys Baker. 

a. Inflorescence pendulous from a long flexible peduncle, glabrous ; rachis 
very flexuose, internodes long, bracts narrow, red ; leaves very glaucous 
beneath. II. Collin si an a. 

b . Inflorescence erect. 

1. Inflorescence long-peduncled. 

(a) Branch -bracts very close together, narrow, truly distichous, hori- 
zontally extended ; rachis very straight. //. librata, 

(b) Branch-bracts distant, not strictly distichous ; rachis very flex- 

uose. 
(i) Inflorescence scarlet, hairy, below the bases of the leaves ; 

leaves green on both sides. //. tortuosa. 

(2). Inflorescence orange, glabrous, with a very long peduncle 

which raises it up among the leaves ; leaves slightly glaucous 

below. //. latispatha Benth. 

2. Inflorescence sessile; rachis straight; branch-bracts narrow, distant, 
red and yellow, covered all over with short velvety pubescence. 

//. spissa . 

3. Inflorescence sessile or very short-peduncled ; branch -bracts very 

broad and close together, often clasping each other at the bases 
and concealing the stiff, straight rachis. 

(a) Branch-bracts green with red cheeks; leaves green on both 
sides when well developed, very long, long-cuneate at the base. 

II. elo?igata. 

(b) Branch-bracts bright orange; leaves glaucous below, at least 

when young, shorter, not cuneate. 

(1) Stem upright; leaves turned in all directions, soon green 
beneath ; tips of the branch-bracts acute or bluntish, some- 
what incurved and involute ; Guatemalan. //. Champneiana. 

(2) Stem inclined; leaves all spread out horizontally, very 
glaucous beneath ; tips of the branch-bracts long- acuminate, 
recurved and revolute ; Porto Rican. //. Borinquena. 

B. Branch-bracts very close together on the rachis so as to form a compact flat 
spike, somewhat resembling the rattle of a rattlesnake (H. imbricata). 

Subgenus Taeniostrobi7s Kuntze. 
Inflorescence pendulous on a long peduncle ; plant very large. 

H. Maria e Hook. f. 

It will be noticed that in this key the species are arranged in a 
considerably different way from that used by former writers. 
Baker and Schumann divide the genus into two subgenera charac- 
terized by broad and narrow branch-bracts respectively. Under 



644 Griggs : Species of Heliconia 

these subgenera the species are arranged according to an artificial 
key. This arrangement often separates widely species which are 
very similar except in some unimportant character, and thus ignores 
the existence of natural groups smaller than the subgenus. That 
such groups do exist is certain, e. g., that referred to under H. 
Champneiana and that to which H. tortuosa and H. latispatha be- 
long. It is believed that their recognition will be of material 
assistance in the study of the genus. The delimitation of the sub- 
genera has also been considerably emended. It is not to be 
denied that the shape of the branch-bracts is in a general way corre- 
lated with the relationships of the species, but it is only an acci- 
dental parallelism without much physiological importance, for 
there are many exceptions — species quite similar in all respects 
except that the branch-bracts are sufficiently different to place 
them iii different subgenera as heretofore defined ; for instance, H. 
tortuosa and H. latispatha. It is an undesirable character to use 
in a key because of the difficulty of application and the liability to 
error. In the inflorescences of several species the lower branch- 
bracts are lanceolate, while the upper are ovate and the intermedi- 
ate neither one or the other. Besides, there are several species 
whose branch-bracts are ovate when fresh, but shrink so as to be 
lanceolate in the herbarium specimen. 

Two of the subgenera, Stenochlamys and Taeniostrobus, are 
natural groups, but the third, Platychlamys, the name of which 
must of course go with the type, comprises simply the residue. 
Its present application can be considered only provisional. The 
extreme form of Stenochlamys (H. psiitacorum) is very different 
from the members of the other two groups, but the difference is 
of only subgeneric importance, for these extremes shade through 
a series of intermediates into the other groups. This is true 
whatever characters are used to distinguish them, whether the 
width of the bracts or the habit, but the habit represents a natu- 
ral difference of some physiological significance ; and it seems to 
me to accord with the probable relationship of the plants. 

In the groups with the typical Heliconia habit, those species 
with erect inflorescences have been separated from those whose 
inflorescences droop. The importance of this character seems 
not to have been appreciated heretofore because of the lack of 



Griggs: Species of Heliconia (>45 

field study, and the descriptions of more than one species would 
have them with indifferently erect or drooping panicles. But it 
seems very probable that such inferences are the result of identi- 
fying two species as one. The ecological relations of the plants 
in the two cases are quite different. The bracts in the erect inflo- 
resence are always full of water, into which numerous insects fall 
and decay, and in species like H. Borinquena the floral parts them- 
selves are always more or less rotten. The fruit of such species 
ripens under water ; and both flowers and fruit have the full bene- 
fit of the sun. But in the pendent inflorescence the flowers are 
shaded from the sun and rain by the roof-like bract above them, 
and are subjected to very different conditions. It is a question 
whether the erect-flowering species are insectivorous or not. If 
they are not, they would have only to utilize a food supply 
already present to become so. That this habit is no mere acciden- 
tal peculiarity of the individual plant will be realized at once by 
any one who sees in the field how constantly the species adhere to 
it. With very rare exceptions the erect-flowering species always 
bear the inflorescence erect, no matter what may be the position 
of the stem, and vice versa. It is as constant as any character 
they possess. 

Another character which has not received the attention it de- 
serves is color in the inflorescence. The neglect of this is due 
largely to the method of dealing with dried specimens from which 
it is absent. My own observation in the field goes to show that 
it is one of the least variable characteristics of the genus. The 
shape of the parts, their relative size, and the presence of hair 
and glaucescence — these are variable and must be used carefully. 
But in the study of several thousand individuals of the more 
abundant species no variation in color was detected. The colors 
of the inflorescence are nearly always bright, but never delicate 
or such as are usually subject to variations. 

Heliconia Choconiana S. Watson. 

Whole plant glabrous, slender, as much as 2 m. tall : leaves 
reaching a length of 33 cm. and a breadth of 7 cm., oblong, acute 
or short-acuminate at the apex, cordate and almost clasping the 
stem, green on both sides : peduncle sometimes as much as 7 cm. 



(346 Griggs : Species of Heliconia 

long, rachis flexuose, deflexed ; lowest fertile bract 7 cm. long, 
2 cm. broad : pedicels 1 2 mm. long : berries red and yellow, 9 
mm. in diameter in the dry specimen. 

. Our plants {110. 757) were larger than Watson's and the pe- 
duncle is longer ; otherwise they are very similar. They were 
growing in deep shady forests at Cherujija Oxec, about fifty miles 
west of the original locality. 

Donnell Smith's no. 8019, collected by von Turckheim at Cubil- 
quitz, Alta Vera Paz, has a peduncle as long as our plants. From 
this it would seem that Schumann's distinction (a sessile inflores- 
cence) between H. aurantiaca and H. Choconiana will not hold. 
H. Choconiana is very close to H. aurantiaca, or at least to H. 
brevispatha, which was described as synonymous with it. The 
peduncles of our specimens are as long as those shown in Hook- 
er's figure and the leaves are similar, but in his species they are 
broader with short petioles, not clasping. The distinction between 
the two should be the floral and foliar characters instead of the 
length of the peduncle. 

Heliconia crassa sp. nov. 

The habit of this plant may be said to be somewhat interme- 
diate between the typical Stenochlamys and the rest of the genus. 
The stem elongates between the nodes so that the leaves are borne 
along it, as is the case in the former group ; but the petioles, in- 
stead of diverging at an angle from the stem, continue in nearly 
the same direction, which is characteristic of the other subgenera. 
They have at the summit, however, a sharp bend of nearly 90 , 
so as to bring the blades into the same relation to the stem as is 
usual in Stenochlamys. Such a bend at the summit of the petioles 
was also seen in several species of the other subgenera (only 
among those with erect stems); but its amount was small, in no 
case more than 30 . When growing in shady places this species 
may attain a height of 2 m., but most of the plants are only about 
half as tall. 

Leaves to 30 cm. long, 9 cm. broad, lanceolate, at the tip 
acute, at the base oblique or equilateral, rounded or cordate, 
thick, green, glabrous, and glossy on both sides ; principal nerves 
on the upper surface 5 mm. apart — an unusual distance for so 
small a leaf; blades sessile or on petioles up to their own length: 



Griggs : Species of Heliconia 647 

inflorescence glabrous ; peduncle 5-30 cm. long, pointing in the 
same direction as the stem, not necessarily upright ; panicle about 
7 cm. long, of half a dozen horizontally divaricate bracts on a flex- 
uose vertical rachis, orange-yellow shading into orange-red on the 
cheeks of the bracts ; the lowest bract often sterile, ascending 
and expanded into a greater or lesser blade, the others distant 
from each other by about their own depth, 3-6 cm. long, about 2 
cm. wide, mostly subovate : flowers about 15 to a branch-bract, 
yellow, as long as the upper bracts : flower-bracts shorter than 
the flowers, 2-3.5 cm - l° n g» nearly 1 cm. broad : pedicels about 1 
cm. long : berries 5 mm. in diameter in the dry specimen. 

Heliconia cr asset, is abundant on the mountains between Sepa- 
cuite and Secanquim (on the road from Cahabon to Senaju), but 
does not cross the divide into Sepacuite. It is oftenest found on 
the hot hills of the region, which are overgrown with bushes and 
small trees. It prefers the full glare of the sun. Nos. j$6 and 
376 (type). 

It is close to H subulata Ruiz & Pavon ; it resembles their 
figure closely, but on examination of the description it is found to 
differ in several respects. It is only a third as large as H. subulata 
sometimes grows ; the leaves of that species reach a meter in 
length and the inflorescences 30 cm.; the panicle in H. crassa is 
erect, not pendulous ; the branch-bracts are much broader, not 
yellow, tipped with red and not turning red in fruit ; and the flowers 
are fewer to a bract. Its similarity to H subulata suggests possible 
affinities with the other forms referred to H. psittacorum L. as varie- 
ties. It has no resemblance to H. angusta Veil., which has been 
considered a synonym of H. subulata ; beside differences in the in- 
florescence, that species has leaves more than eight times as long 
as broad. Heliconia Andrewsii KL, which has been considered 
another synonym of H. subulata, is nearer, but it also has longer 
leaves ; its spathes are narrower, not channelled, and its flowers 
are very much larger, orange, and distinctly black-tipped. From 
H. Schomburgkiana KL, which, however, is very inadequately de- 
scribed, it differs at least in the broader conduplicate branch-bracts. 
Eichler's variety, spathacea, which has been considered identical 
with it, has, as shown by Petersen's figure, long narrow leaves 
almost like the species. Petersen's variety, gracilis \ has not been 
taken up by recent writers because of insufficient description ; the 



648 Griggs : Species of Heliconia 

branch-bracts are broad like those of H. crassa, but no mention is 
made of the leaves, which would, therefore, be expected to be like 
those of H. psittacorwn and very much narrower than those of the 
present species. 

Heliconia Collinsiana sp. nov. 

The few plants of this species we saw were growing in a clump 
at the top of a steep bank by the roadside. They were lopping 
over upon the surrounding shrubbery as though unable to stand 
alone ; the leaves were spread out to catch the sun like those of 
H. Borinqnena (/. j). 

Whole plant about 5 m. long : leaf -bases 2.75 m., petiole 1 m., 
blade 1.25 m.; blade oblong, gradually narrowed to the tip, ob- 
liquely cordate at the base, decidedly inequilateral in the type, the 
sides being 17 and 22 cm. respectively, bright green above, below 
very glaucous, with a little fine fuzzy pubescence on the midrib, 
otherwise glabrous : inflorescence bright red, glabrous, 45 cm. 
long, pendent on a slender peduncle 30 cm. long ; branch-bracts 
lanceolate, the lowest linear-lanceolate, gradually narrowed to a 
blunt tip, very distant on the very flexible but slightly flexuose 
rachis, lowest 30 cm. long, 4 cm. broad, those half way up 15 
cm. long and 3 cm. broad : flowers and their bracts not seen ; few 
flowers (less than a dozen) to a branch-bract: pedicels 15-25 
mm. long : berries 1 5 mm. in diameter when fresh. 

On the bank of the Cahabon river about five miles below the 
ford by which the road crosses to the town of Cahabon {110. J52). 
It gives me great pleasure to name this species after my friend and 
associate, Mr. G. N. Collins ; but for his pointing it out to me I 
would have ridden by without seeing it. 

Donnell Smith's no. 2102 from Masagua, Escuintla, is probably 
the same species. The leaf with his specimen is only 52 cm. 
long, 25 cm. wide, ovate and green below ; its slender petiole, 
however, shows that it is not one of the large leaves of the plant, 
but from one of the small lateral suckers whose leaves are always 
shorter than those of the main stem. The same fact may also 
explain the absence of glaucescence, but the whole group is very 
variable in that respect. The inflorescence of this specimen re- 
sembles that of the type closely, but the branch-bracts are some- 
what broader and each has more flowers. 

According to the artificial key given by Schumann Heliconia 
Collinsiana could belong only with H. piilvertdenta or H. glanca, 



Griggs : Species of Heliconia 649 

but it is very different from both. H. glauca has a red rachis, 
pedicels and fruit, combined with green bracts and flowers. H. 
pidveridenta belongs to Stenochlamys ; its peduncle is not droop- 
ing, the inflorescence is much smaller with fewer bracts, the rachis 
straighter and the bracts closer. The relationships of the present 
species are more probably with H. pendida. Donnell Smith's no. 
4635 '> which was doubtfully labelled as that species by Baker, is 
closely related to the present plant but has a very hairy inflores- 
cence and differs in other characters. It seems, however, to dif- 
fer from H. pendula also. The present species differs obviously 
from the latter in the glabrous inflorescence and the longer branch- 
bracts. 

Heliconia librata sp. no v. 

At the time of our visit this species had passed its season and, 
though many plants were observed, none were seen with fresh 
flowers ; the inflorescences of all were dead and dry like the type, 
which was in the same condition when collected as now. The 
habit of the plant is like that of H. latispatha, which the present 
plant resembles much in all respects except the inflorescence. It 
is erect in all its parts. 

About 3 m. tall : stem 75 cm. long : petioles 90 cm. long ; 
blade 115 cm. x 25 cm., oblong-elliptical, rounded and abruptly 
short-acuminate at the base, acute or slightly acuminate at the tip, 
glabrous, glaucous below : inflorescence on a very long (75 cm.) 
erect peduncle, green according to the statement of the Indians, 
triangular in outline, almost as broad as long, composed of 12-16 
bracts about a centimeter apart, most of them extended at right 
angles to the straight rachis ; type inflorescence glabrous except 
for a few hairs on the rachis ; branch-bracts (except the upper) 
lanceolate, the lowest sterile, 20 cm. long, 4 cm. broad, with 
a small lamina at its tip ; largest of the fertile bracts about half 
as long but of the same width, the topmost only 3 cm. long : pedi- 
cels 10-15 cm. long. 

Heliconia librata is a lowland form, abundant in the upper part 
of the Oxec valley, but does not occur in any other region we 
visited (no.. 696, Cherujija Oxec). 

According to the artificial keys it belongs in Baker's section 
Stenochlamys and would be nearest to H. glauca, which it scarcely 
resembles even in general appearance and differs from sufficiently, 
as far as characters go, in having the leaves more than twice as 



650 Griggs : Species of Heliconia 

large, the internodes of the rachis shorter and the bracts more 
numerous. The relatively short rachis and the long peduncle, 
with the comparatively large number of branch-bracts, suggest 
that this plant is near those species with a head of densely imbri- 
cated bracts for which the section Taeniostrobus was erected. 

Heliconia tortuosa sp. nov. 

Whole plant about 3.5 meters in height (trunk 1.75 m., petioles 
75 cm.) : leaves about a meter long, 20-30 cm. broad, rounded 
and oblique at the base, abruptly short-acuminate at the tip, bright 
green and glabrous on both sides except that the midrib, below, 
and the petiole bear more or less coarse matted brown hair : in- 
florescence brilliant scarlet, erect or nearly so ; peduncle long, 
stiff, aligned with the stem ; lowest bract often developing a large 
blade, sometimes nearly as large as the other leaves ; rachis 
extremely flexuose ; the few bracts (about half a dozen) distant 
from each other by about their own depth, not truly distichous 
but arranged in a sort of spiral with an angle of about 120 be- 
tween them, lowest about 12.5 cm. long, 5 cm. broad ; the upper 
nearly as large, subovate, triangular, straight-sided, not tapering, 
with a blunt frayed point ; rachis between the upper bracts attain- 
ing its full extension almost as soon as they separate from the 
head formed by those above them which have not yet opened out ; 
bases of bracts at least, and rachis covered with matted brown 
wool : flowers green, 5-7 cm. long, projecting above the edges of 
the branch-bracts ; flower-bracts two thirds as long as the flowers ; 
floral parts glabrous except for an occasional hair. (Pl. 29, f. i.) 

No. 17 ; photographs 3610 and 4130. Heliconia tortuosais the 
only representative of the genus common about Sepacuite, but in 
the valley which the plantation occupies it is one of the most con- 
spicuous plants ; it grows everywhere except in the thickest 
woods where the light is too weak. Donnell Smith, no. 1828, 
Pansamala, Alta Vera Paz, alt. 3800 feet, April, 1899. The bracts 
are longer and narrower than in our specimen. 

Heliconia tortuosa is very similar to H. villosa Kl., from Colom- 
bia, or at least to the printed descriptions of it. The similarity is 
due, I suspect, more to lack of detail in the descriptions than to 
resemblance in the plants. Petersen's plate in the " Flora Brasi- 
liensis " shows a very different plant from the present and empha- 
sizes the discrepancies between the two. The Guatemalan species 
differs from the Colombian in having (1) a peduncle straight and 



Griggs: Species of Heliconia 651 

stiff, not curved ; (2) an oblong, not a deltoid, panicle ; (3) a very 
flexuose rachis ; (4) bracts not approximated and fewer than in 
H. villosa ; (5) flower-bracts glabrous, shorter than the flowers, not 
hirsute and of the same length ; (6) flowers larger, glabrous, not 
hairy, pedicelled, not sessile, ovary glabrous, not pubescent ; (7) 
leaves obliquely rounded at the base and short-acuminate at the 
tip, not subattenuate, alike at both ends. 




Fig. 1. Heliconia latispatha Bentham. Inflorescence just before anthesis. Note 
that though the branch-bracts are quite narrow, the flowers are not visible. Traced 
from a photograph by Mr. G. N. Collins («<?. i2g). 

Heliconia latispatha Bentham 

Whole plant about 2.5 m. tall ; stem 15 cm. in circumference, 
green at the base but above and on the petioles covered with a 
thin evanescent bloom : petioles about 50 cm. long : leaf- blades 
1 m. by 30 cm., short-acuminate at the tip, obliquely narrowed to 



652 Griggs: Species of Heliconia 

the base, glabrous on both sides except for the midrib which is 
somewhat hairy, slightly glaucous below : inflorescence borne at 
the summit of a long (60 cm.) peduncle ; lowest bract, which is 
sterile, and sometimes the next, expanded into leaves which may 
attain a large size; branch-bracts 8-10, rich orange, rachis green- 
ish-orange ; bracts narrowly lanceolate ; one from the middle is 
1 5 cm. long, 3-4 cm. broad (dry) : flowers very small, entirely 
hidden below the edges of the bracts, 30-35 mm. long, subtended 
by triangular-ovate bracts of about the same length and 12-15 
mm. broad, pedicels apparently not pubescent as they are described 
in H. latispatha and there is no hair on any part of the inflores- 
cence. (Fig. 1.) 

Secanquim, about half way between Senaju and Cahabon (110. 
262, photographs 128, 129). It is very abundant on the hillsides 
around Secanquim, where it occupies the same place in the plant 
societies as does H. tortuosa higher up about Sepacuite. 

Donnell Smith, no. 1829, Rio Dulce, Dept. Livingston. In 
this specimen, which is more mature than ours, even good sized 
fruits are very short-pedicelled and barely look over the edges of 
the narrow branch-bracts. 

The panicle is oblong in outline like that of H. tortuosa, to 
which the present plant is closely related. As in that species the 
branch-bracts are not truly distichous, but are only approximately 
opposite and the rachis is strongly flexuose. But the peduncle is 
much longer, so that the inflorescence though originating at about 
the same part of the plant is much higher up ; instead of being 
half way down the stem it is about level with the bases of the 
leaves. 

Heliconia spissa sp. nov. 

Plant about 2 m. long, with a habit like H. Borinquena (f. 3), 
leaning over with its leaves all horizontal : blade 75 cm. long, 20 cm. 
wide, oblong-oval, acute at the tip, rounded to the base, green and 
glabrous except for the under side of the midrib which bears more 
or less coarse brown hair; main nerves only 3 mm. apart — 
whence the name : inflorescence practically sessile (on a pe- 
duncle only 25 mm. long), stiffly erect, thickly covered with soft 
short brown hair except on the edges and channels of the bracts, 
rachis straight, hardly flexuose ; branch-bracts distant from each 
other by more than their own depth except near the tip of the 
panicle, oblong-lanceolate, acute or obtuse, the lowest 17 cm. 
long, 1.5 cm. broad; lowest fertile bract 8 cm. long, 1.5 cm. 



Griggs : Species of Heliconia 653 

broad, uppermost 4 cm.; whole inflorescence bright yellow, shad- 
ing to orange on the cheeks of the branch-bracts : flowers a dozen 
or less to the bract, about 4 cm. long, yellow, densely covered 
with the same soft short hair as the rest of the inflorescence, not 
scabrous; flower-bracts 25 mm. long, nearly 10 mm. broad, tri- 
angular acute, densely pubescent, especially on the midrib ; pedi- 
cels 1 cm. long, hairy : berries (not ripe) also hairy. 

Along the road from Cahabon to Senaju (no. 359) growing on 
a steep river bank a mile further east than 77. Collinsiana. It is 
known to me from a single clump with only one inflorescence. 

Heliconia spissa very nearly corresponds with the descriptions 
of H. Schiedeana. But they are lacking in detail except as to 
floral characters. The most obvious difference is in the color of 
the branch-bracts, which is red in that species but yellow and 
orange in the present. 

Heliconia elongata sp. nov. 

Plant generally erect in all its parts, 2-4 m. tall (type, 3.5 m, 
with a stem 1.5 m., petiole 0.5 m., blade 1.5 m.), glabrous: leaf 
elongated-oblong, only one seventh as broad (22 cm.) as long, 
very long-cuneate at the base, rounded to the suddenly short- 
acuminate tip, green on both sides ; mid-rib very large and stiff, 
1 5 mm. in diameter near the base of the leaf : inflorescence up- 
right, without a peduncle, oblong in outline, composed of a dozen 
or more deeply boat-shaped bracts with clasping bases which 
hide the stiff, straight rachis ; branch-bracts green on the edges 
and bottom, shading through yellowish-green to a small patch of 
weak, light red on the cheeks but nowhere crimson ; one from the 
middle of the spike 13 cm. by 9 cm. (measured dry), all ovate, 
gradually tapering to the sharply acuminate tip : flowers 8 cm. 
long, bright green at the exposed tips, just projecting above the 
edges of the branch-bracts ; segments of the perianth linear ob- 
long ; flower-bracts 5 cm. long, 2 cm. broad, triangular-ovate. 
(Fig. 2.) 

No. 790. It grows all along the Polochic River and up into 
the hills to an altitude of about 2,000 feet along the road from 
Panzos to Sepacuite, where a few clumps of it were seen growing 
with H. Champneiana. It covers a greater area and thrives under 
more varied conditions than any other species studied. 

Donnell Smith, no. 1830, " Monte Cachirulo, Depart. Yzabal, 
alt. 700 pp. April, 1899." 



654 



Griggs: Species of Heliconia 



Heliconia clongata is closest to H. humilis but differs in stature, 
in the shape of the leaves and in having nearly twice as many 
branch-bracts, which are not nearly as highly colored, those of 
H. humilis being deep scarlet with only a narrow margin of green 
near the tip. It also resembles H. nttila (see below), but differs 




Fig. 2. Heliconia elongata sp. nov. The type clump, showing how conspic- 
uously erect are the inflorescences, regardless of the stem angle. Traced from a photo- 
graph. 

in the narrow perianth-segment, the wider more numerous green 
and red branch-bracts and the shape of the leaves. 

In studying Heliconia elongata, it has been necessary to deter- 
mine H. Bihai. That species is the type of the genus and for a 
long time was the only one recognized. As is not unusual in 
monotypic genera, to the one species were referred all the Heli- 
conias discovered for a long time. In this way H. Bihai became 



Griggs: Species of Heliconia 655 

a ''composite" to which different authors referred very different 
plants, and consequently has been a source of much difficulty and 
confusion. This confusion has continued to the present time, and 
the recent monographers have not cleared it away but have made 
their descriptions so general as to include most, though not all, of 
the various plants described in the places they cite. To determine 
what H. Bihai really is and to limit the name to a single species 
is not an easy task, but it is necessary for any precision in dealing 
with it and the related species. In attempting to make such a de- 
termination we are aware of the great possibility of error, but we 
consider it less serious to commit a nomenclatorial blunder by ap- 
plying the name to the wrong plant than to make the taxonomic 
mistake of referring several species to one. Besides it is more 
likely that any nomenclatorial uncertainty will be removed, if it 
can be, after the species themselves are differentiated. 

For some reason Linnaeus in what is, for nomenclatorial pur- 
poses, the original description * in the " Species Plantarum," cited 
Plumier's last species {variegatd) first. Were this determinable it 
would be the type of H. Bihai, but beyond the reference to the 
variegated branch-bracts we know nothing about it and there 
seems to be no way of finding out anything else. Therefore we 
reject this species from consideration and take up the next cited, 
one with scarlet bracts, which was Plumier's first species. The 
early French writers that had access to Plumier's manuscripts are 
united in considering that this was the species figured by Plumier 
and in calling it Heliconia Bihai. Their figures and descriptions 
are in substantial accord with his but add a great deal to them. 
Burmann, in 1756, published a plate from Plumier's manuscripts 
showing the whole plant and the floral parts full size. The draw- 
ings of the flowers are almost identical with those published 
earlier by Plumier himself, and almost certainly came from the 
same species. Plumier, in the earlier work, gave no hint as to 
which of the three species he was figuring, but the identity of his 

* " Musa spadice erecto. Bihai. 

Bihai amplissimis foliis, florum vasculis variegatis. Plum. gen. 50. 

Bihai. Ovid. I. 8, c. 9. [Should be lib. 7. cap. 9. — R. F. G.] 
/?. Bihai amplissimis foliis, florum vasculis coccineis. Plum. gen. 50. 
y. Bihai amplissimis foliis, florum vasculis subnigris. Plum. gen. 50. 

Habitat in America calidiore" — L. Sp. PI. 1043. 1753. 



656 Griggs : Species of Heliconia 

figures and Burmann's show that both were drawn from the spe- 
cies with scarlet branch-bracts which alone was referred to by Bur- 
mann. This Heliconia Bihai may be briefly described as follows : 

Heliconia Bihai L. 

Bihai florum vascidis cocciizeis Plum. N. PL Am. Gen. 50. pi. J. 
1703. 

Musa Bihai L. Sp. PI. 1043. 1 7S3 ( m part). 

Bihai Plum. PI. Am. ed. Burm. 49. pi. 59. 1756. 

Heliconia Bihai L. Mant. PI. 211. 1767 (in part). Jacq. PI. 
Rar. Hort. Schoen. 1 : 26. 1797. Lam. Encyc. pi. 148. 1823. 
L. C. Richard, Nova Acta, 15, suppl. : 22. pi. 8, 10. 1831. 

Whole plant 3-4 m. tall, erect, glabrous : leaves subdistichous, 
nearly 2 m. long, 50 cm. broad, round at both ends, long-petioled : 
inflorescence stiffly erect ; peduncle thick, as long or longer than 
the straight rachis, raising the inflorescence up among the leaves ; 
branch-bracts scarlet, broadly ovate, acuminate, mostly concealing 
the rachis : perianth-segments narrow, oblong linear ; flower-bracts 
ovate, acute. 

The habitat of this plant is uncertain, but it probably came 
either from one of the French West Indies, most likely Martinique, 
or from Guiana. 

This clears the way for the consideration of the other plants 
confused with H. Bihai. The first was described by Swartz (Obs. 
Bot. 96. 1 791). The description is mostly of flowers and is 
hardly definite enough for specific determination. Its bracts, 
however, are different in color from those of H. Bihai and on that 
account it was renamed H. lute of use a by Jacquin. 

In Andrews's Repository [pi. 64.0) is figured as H. Bihai a 
species with a sessile inflorescence whose bracts are purple with 
bright yellow edges. In general appearance it is similar to H. 
humilis, though taller. For this the name Heliconia purpurea is 
suggested. 

Another so-called Heliconia Bihai was figured by Loddiges 
(Bot. Cab. pi. 252) and by Edwards (Bot. Reg. pi. 374). The 
plant is about 3 meters high, with few leaves and erect habit like 
H. elongata. Inflorescence sessile, upright ; rachis red (visible), 
but little flexuose ; bracts narrow, ovate-lanceolate, hardly touch- 
ing each other, red with yellow margins ; flower broad and short, 



Griggs: Species of Heliconia 657 

perianth-segments broad, flower-bracts ovate. It differs from H. 
Bihai in the habit, in the acute leaf and in the sessile inflorescence ; 
and from H. purpurea in the color and width of the bracts. For 
this plant the name Heliconia rutila is suggested. 

A fourth species is described by Petersen in the Flora Brasili- 
ensis (3 3 : 16. pi. 5). This, Petersen himself believed to be dif- 
ferent from Richard's species, but he supposed it to be the same 
as Swartz's, to whom he credits the name. It differs from H. 
Bihai in having acuminate glaucous leaves ; a long, weak, flexible 
peduncle, a very flexuose rachis with intern odes 3-5 cm. long ; 
distant, narrow, branch-bracts ; and triangular flower bracts. For 
this species the name Heliconia distans is proposed. As the Flora 
Brasiliensis is not accessible to everyone, it may be of interest to 
add that a reduction of Petersen's plate of H. distans is given in 
Schimper's Pflanzen-geographie, page 359. 

For a summary, a key to the species confounded with Heli- 
conia Bihai may be given : 

Peduncle long, stiff, erect ; branch-bracts red, broad, close together ; leaves round at 

both ends, green on both sides. II. Bihai. 

Peduncle long, flexible, curved ; branch-bracts red and yellow, narrow, distant ; 

leaves acuminate, glaucous below. II. distans. 

Peduncle none, leaves green on both sides, acute. 

Branch-bracts red and yellow, narrow, barely touching each other; perianth-seg- 
ments short and broad. II. rutila. 
Branch-bracts purple and yellow, broadly ovate, close together ; perianth-segments 
linear. //. purptirea. 

Heliconia Champneiana sp. nov. 

Whole plant 5.5 m. tall (stem 1.5 m., petiole 1.5 m., blade 2 
m.), erect : stems, petioles and blades of the leaves upright and 
turned in all directions, i. e. y tangential to circles centering in the 
stem : stem beset with long, straight light brown hairs projecting 
at right angles : leaves 2 m. long, about 50 cm. wide, tip rounded, 
not acute, base obliquely cordate ; covered when young with a 
slight bloom on the under surface, which disappears with age and 
in drying, leaving them green and glabrous on both sides : in- 
florescence about 45 cm. long, oblong in outline, composed of 
about 9 bracts, bright orange with occasional red splashes : pedun- 
cle short and stout, not more that 25 cm. long, hairy; branch- 
bracts close together, hiding the rachis and touching the base of 
the next above them, on the other side ; upper nearly uniform in 
shape and size, about 14 x 11 cm. (measured dry), broadly ovate, 



658 Griggs : Species of Heliconia 

deeply boat-shaped, slightly tapering to a blunt somewhat in- 
curved tip ; rachis straight, glabrous like the bracts ; each bract 
very many-flowered: flowers about 7.5 cm. long, green outside, 
white inside, pubescent in two broad lines up the back, otherwise 
glabrous ; flower-bracts glabrous, nearly as long as the flowers, 
quite broad. (Pl. 30.) 

No. 528 ; photographs {, j. Helicouia Champneiana was first 
seen on the steep mountain side up which zigzags Mr. Champ- 
ney's road from Panzos to Sepacuite. Later it was observed east 
of Cahabon in the valley of the Oxec river. It gives me much 
pleasure to associate Avith this species the name of Mr. Kensett 
Champney, whose great and precise knowledge of the flora, fauna, 
customs and language of the country makes him an authority on 
all that pertains to the natural history and anthropology of Alta 
Vera Paz. To him we are indebted for a great deal of valuable 
information and for all the courtesies which he and Mr. and Mrs. 
Owen united to show us during our visit to Sepacuite. 

Heliconia Champneiana may be taken as the type of a very 
compact and natural section of the genus including H. rittila, H. 
purpurea, H. elongata and H. Borinquena. These are all character- 
ized by practically sessile, erect inflorescences, stiff straight rachises 
and by relatively broad often clasping branch-bracts. 

Heliconia Borinquena sp. nov. 

This species is very similar to the Guatemalan plant just de- 
scribed, but is somewhat smaller in all its parts. Whole plant 
3-4 m. long with 3-5 leaves on petioles about 1 m. long and 
sheathing bases twice as long. Its habit is different from H. 
Champneiana ; instead of growing erect it leans over and spreads 
out all its leaves horizontally. (Fig. 3.) 

Leaves 1-1.25 m. long, about 30 cm. broad, rounded at the 
base, acute or short-acuminate at the tip, glabrous, green above, 
decidedly glaucous beneath : inflorescence conspicuously upright 
no matter what may be the position of the stem, glabrous through- 
out, oblong, about 30 cm. long, half as wide ; branch-bracts 8- 
10, very broadly ovate, broader than in H. Champneiana , all ex- 
cept the lower nearly as broad as long (one from the middle of 
the spike measures 1 2 cm. both in breadth and length), closely 
clasping and overlapping each other at the bases so that the side 
of one is on a level with the bottom of the next above it, on the 



Griggs: Species of Heliconia 



659 



same side of the rachis, much closer than in H. Champneiana ; 
their tips long, subulate, recurved with revolute margins ; flowers 
greenish, about I i cm. long, tips projecting above the edges of 
the branch-bracts ; flower-bracts about 9 cm. long. (Pl. 29, f. 2.) 

Growing on the steep sides of a wet clay ravine amid a tangle 
of tropical underbush, near the military road, 13 km. north of 
Cayey, Porto Rico (Underwood and Griggs no. j6j, June 24, 1901 ; 
photographs 2827 and 28 2 g.} 

Our specimens of H. Borinquena are, with one exception, the 
only representatives of the genus in the National Herbarium from 




Fig. 3. Heliconia Borinquena sp. nov. Whole plant, showing the habit and 
the upright inflorescence. Traced from a photograph by Mr. G. N. Collins (no. 2827). 

Porto Rico. The exception is Sintenis's no. iiji from near Baya- 
mon, which was determined as H. Bihai. Since, however, that 
species has leaves green on the under side such a reference is 
probably a mistake, for the leaves of Sintenis's specimen are very 
glaucous below. Its leaves suggest H. Borinquena, for they are 
of about the same texture and size — as far as one can tell from 
the fragments preserved — but it has a long slender peduncle and 
the bracts are longer, narrower and triangular-ovate, without the 
subulate recurved tip (in the specimen). 



660 Griggs: Species of Heliconia 

Heliconia Borinquena and H. Champ neiana y which is closely, 
related to it, are easily distinguishable from the others of the group 
to which they belong by the orange branch-bracts, for all the 
others are margined with a color different from that of the cheeks. 

Heliconia Mariae Hook. f. 

This species is a giant in a genus of giants ; it sometimes 
reaches a height of 1 2 meters when growing in the shade and 
protected from the weather. Most plants are, however, not more 
than two thirds as large. Its general appearance when growing 
in masses is strikingly similar to that of a banana plantation, 
though the individual plant would never, even if not in flower, be 
mistaken for a banana, because of the smaller number of leaves. 

Leaves 2-3 m. long, about 60 cm. broad, oblong-elliptical, 
obliquely subcordate at the base, acute at the tip, when young 
covered with a slight evanescent bloom : inflorescence a dark rose- 
color, at the end of a long (60 cm.) nodding peduncle; bracts 
densely imbricated, concealing the rachis in the fresh specimen, 
panicle generally not much longer than broad, but sometimes very 
long, with very many branch-bracts ; lowest fertile branch-bracts 
slightly reflexed, ovate, 8-9 cm. long, 7-8 cm. broad, attenuate 
to a blunt frayed tip, narrowed to the base, those from the middle 
of the inflorescence nearly orbicular when spread out, 6-7 cm. 
long, 7-8 cm. broad, widest at the middle, narrowed both ways 
to the blunt tip and base, with more or less short soft brown hair : 
flowers about ten to a branch-bract, their exposed tips rose, fading 
into white on the shielded portions, 4-5 cm. long, with flower- 
bracts a little shorter and sometimes at least 2 cm. broad, often 
conduplicate and keeled below where compressed by the surround- 
ing flowers, part of them hairy; ovary white, turning deep blue 
when ripe ; pedicel very variable in length. 

There are some discrepancies between this plant and the de- 
scriptions of Hooker's plant, but they are mostly in size and hairi- 
ness. In more constant particulars they seem to correspond very 
well. 

This species grows abundantly around and a little above 
Panzos {no. 787). None of it was seen along the Polochic, nor 
does it ascend to Sepacuite. A solitary clump was seen in the 
valley of the Oxec. 

Beside the species already described one other was collected 
in the valley of the Oxec. Only one plant was seen and that 



Griggs: Species of Heliconia <)G1 

toward night when we were hurrying to reach our destination, 
and a part of this specimen was lost, so that it is better to pass it 
till another time when fuller material can be had. It was of the 
subgenus Stenochlamys, perhaps closest to H. cannoidea Rich., 
1.5 m. tall, with seven leaves, 40 cm. long, not quite 10 cm. 
broad, sessile with long acute bases, acuminate at the tip ; green 
and glabrous on both sides; peduncle a little less than 10 cm. 
long ; inflorescence bright scarlet, about the same shade as in //. 
tortitosa, rachis deflexed, flexuose, lowest bract 17 cm. long with 
a small blade at the end, next 9 cm., both about 2 cm. broad (no. 

693)- 

Explanation of Plates 

Plait. 29 

Fig. 1. Heliconia tortuosa sp. now The inflorescence, one half natural size. 
Photographed in the field by Mr. G. N. Collins (no. 3610). 

Fk;. 2. Heliconia Borinquena sp. nov. The inflorescence, one half natural si^e. 
Photographed by Mr. G. N. Collins [no. 2S29). 

Plate 30 
Heliconia Champneiana sp. nov. A portion of the type in situ showing the up- 
right habit. (Photograph no. 5. ) 

LITERATURE CONSULTED 

Andrews, H. Bot. Rep. 2: //. 124. 1800; 10: //. 6^0. 1812. Gives good col- 
ored plates from garden specimens of H. psittacorum Sw. (not L. f . ) (//. 
Andrewsii Kl.) and //. Bihai (//. purpurea), with descriptions of each. 

Attblet, F. Mist. PI. Guiane Fr. 2: 931. 1775. Describes ftlusa Bihai and M. 
hit mi lis. 

Baker, J. G. A Synopsis of the Genera and Species of Museae. Ann. Bot. 7 : 189- 
222. 1893. Describes H. platystachys and //. BurcJiellii. 

Bentham, G. Bot. Sulphur, 170, 171. 1844. The original descriptions of H laii- 
spatha and H. vaginalis ; they are brief, unillustrated, and drawn mostly from vari- 
able floral characters so that they are very unsatisfactory. 

Browne, P. Civ. and Nat. Hist. Jam. 364. 1756. Describes in a general way, as a 
Musa, a Jamaican Heliconia, but does not give sufficient material for specific ref- 
erence. 

Chamisso, A. von & Schlechtendahl, D. F. L. von. Linnaea, 6: 57, 58. 1831. 
A brief collector's note but hardly a description of the plant afterward named //. 
Schiedeana. 

Edwards, S. Bot. Reg. 5 : //. 374. 1819. Gives a good colored plate of //. BiJiai 
(//. rut i la). 



662 Griggs : Species of Heliconia 

Ender, E. Gartenfl. 20: 152. 187 1. A notice of //. vinosa, with a brief descrip- 
tion of it. 

Guilmot, G. Fl. Serr. Jard. 23 : 125. //. 2416-17. 1880. Figures and describes H.? 
striata. It is without flower, and is more likely than not a member of some other 
genus. 

Hooker, J. D. On a new Heliconia with the habits of a Musa> sent from New Granada. 
Jour. Linn. Soc. London, 7 : 68, 69. 1863. Gives a full and complete descrip- 
tion of H. Marine, but no figure. 

Heliconia humilis Jacq. Curt. Bot. Mag. III. 22 : //. 5613. 1866. A 



good description, with a colored figure, of Jacquin's species. 

Hooker, W. J. Heliconia Brasiliensis. Hook. Exot. Fl. 3 : pi. 190. 1827. Orig- 
inal description, full, with a good colored plate. 

■ Heliconia angustifolia* Curt. Bot. Mag. III. 5 : pi. 4473 \ 1849. A good 

description, with a colored plate. 

Heliconia pulveruhnta. Curt. Bot. Mag. III. 7 : pi. 4683. 1852. De- 



scription with colored figure. 

Heliconia metallica. Curt. Bot. Mag. III. 18 : pi. S3 f S- 1862. Original 

description, with colored figure. 

Heliconia brevispatha. Curt. Bot. Mag. III. 19 : pi. 5416. 1 863. De- 



scribed as a synonym of H. aurantiaca Ghiesb. ; a full description illustrated with 
a good figure. 

Jacquin, N. J. PI. Rar. Hort. Schoenb. 1 : 23. pi. 48, 49. 1797. Here is de- 
scribed H. humilis with two large, carefully drawn colored plates. Jacquin is the 
only one of the early writers that had opportunity to study the plants in the field, 
and consequently his descriptions are better and his conception of the relationships, 
number and distinctness of the species are clearer than those of his contemporaries. 

Klotzsch, J. F. Beitrage zu einer Flora der Aequinoctial-Gegenden der neuen Welt : 
Musaceae. Linnaea, 20 : 462-466. 1 847. Describes as new species H. Meri- 
densis, H. villosa, H. Schiedeana, H. Schomburgkiana and H. bicolor, briefly and 
without figures. 

in Schomburgk Reisen Brit. -Guiana, 3: 815, 919, 1070, 1125, 1126. 1848. 



Lists, with notes of flowering, the following : H. acuminata, H. Ballia, H. bi- 
color, H. Bikai, H. Jlexuosa, H. pulverulenta, H. Richardiana and H. Schom- 
burgkiana. 

Kuntze, Otto. Rev. Gen. PI. 2 : 684. 1891. Description of Bihaia imbricata and 
of section Taeniostrobus, with a list of the species, which are transferred to 
Kuntze' s genus Bihaia. 

Lamarck, J. B. A. P. M. Encycl. 1 : 426. 1783; pi. 148. 1823. Describes all 
the Heliconias known at that time, as he understood them, and figures H. Bihai. 

Lindley, J. Bot. Reg. 19 : pi. 1648. 1834. Original description of H. pulverulenta, 
with colored figure. 



Griggs : Species of Heliconia 663 

Linnaeus, C. Sp. PI. 1043. 1753; ed. 2. 1477. I 7^ > 3- First binominal reference 
to a Heliconia under the name Musa Bihai. 

Mant. PL 2ir. 1771. The original application of the generic name. 

Syst. ed. 13. 204. 1774. The first description of the genus. 



(filius) Suppl. Syst. Veg. 157-159. 1781. Describes H alba, H. psitta- 

corum, II. hi r sul a. 

Loddiges, C. Bot. Cab. 3 : pi. 232. 1 81 8. Gives a colored habit-figure of II. Bihai 

(H rtitila). 

Merian, M. S. de. Hist. Gen. Insect. Surinam. 1 : 54. pi. 34. 1 77 1. Gives a de- 
scription with a plate of a Heliconia called " Ballia" by the Indians. 

Miller, P. Gard. Diet. 1739^ A popular and indefinite description of "Bihai." 

Oviedo, G. F. de. Hist. Gen. y Nat. Indias. lib. 7 : cap. 9. Gives an account of a 
plant called " Bihas " and its uses. 

Paxton, J. Heliconia Brasiliensis. Paxt. Mag. Bot. 3 : 193. 1 837. Gives a good 
colored figure and a habit-figure, in addition to a full description. 

Petersen, 0. G. Musaceae, in Mart. FI. Bras. (fasc. 107) 3 3 : 1-27. //. 1-8. 1890. 
Describes 25 species, the new ones being //. elegans, H. conferta, H. Wagneriana, 
II. Bourgaeana and II. cwtispatha. 

Musaceae, in Engler & Prantl, Pflanzenfamilien, 2 6 : 1-10. 1 889. 



Plumier, C. Nov. PI. Am. Gen. 50. pi. 3. 1703. Describes three species of Bihai 
and figures the flower of one. 

PI. Am. ed. Burm. 49. pi. 30. 1756. A better figure and description of 

Plumier' s plant. 

Poeppig, E. Reise in Chile, Peru und auf dem Amazonstrome, 2 : 295. 1836. The 
original application of II. velligera, without much description. 

Regel, E. Gartenfl. 16 : 90, 245. 1867. A review of the original description of 
//. pendnla. — Notice of H. humilis Jacq. 

Richard, L. C. Mus. Coram. Bot., Nova Acta. 15: suppl. 1831. A very impor- 
tant paper illustrated with twelve plates in which //. Bihai, II. acuminata, and II. 
cannoidea are figured. 

Rodigas, E. 111. Hort. 29 : 59. //. 448 : 29 : 155. pi. 464. 1882. Figures and de- 
scribes II ? triumphans and H? aureostriata. Both these are foliage plants with- 
out flowers, and are as likely Cannas as Heliconias, which they do not particu- 
larly resemble in habit. 

Ruiz, H. & Pavon, J. Fl. Per. & Chil. 3: 70, 71. pi 303-303. 1802. Describe 
and figure H. subulata, H. lingulata and H. rostrata. 

Rumphius, G. E. Herb. Amboin. 5: 142. //. 62. f. 2. (1695). 1747. A refer- 
ence to a Musaceous plant, supposed to be a Heliconia, found wild in Amboyna. 



664 Griggs : Species of Heliconia 

Schimper, A. F. W. Prlanzen-geographie. Jena, 1898. On page 359 is given a re- 
duced reproduction of Petersen's figure of H. Bihai (//. distans). 

Schumann, K. Musaceae, in Engler's Pflanzenreich (Heft 1), 4 45 : 1. 1900. 

Sims, J. Heliconia psittacorum Sw. Curt. Bot. Mag. 14: //. 502. 1801. Colored 
plate and description of H. Andrewsii. 

Swartz, 0. Obs. Bot. 96. pi. 5. /. 2. 1791. Describes H. Bihai Sw. and H. 
psittacorum Sw. [H. Andrewsii Kl. ), and figures the former. 

Vellozo, J. M. Fl. Flumin. 3: pi. 19-22. 1827. Text (Archiv. Mus. Nac. Rio 
Jan.), 29. 1 88 1. Figures and describes //. Biahij, H. angttsta, H. t/ialia, H. 
episcopalis. 

Verlot, B. Heliconia glauca. Rev. Hort. 40 : 112. 1869. Heliconia dens, [flora , 
/. c. 274. Very good descriptions accompanied by colored plates. 

Watson, Sereno. Proc. Am. Acad. 23: 284. 1888. Describes//. Choconiana: 

Gard. and For. 1 : 161, f. 31. 1 888. Repeats the description, and gives a 



figure of the plant.